Does anyone have an example of how they have dealt with StaleObjectErrors caused by having optimistic locking enabled?
Ideally I'd like to merge the changes, but I haven't found much in terms of sample code or documentation on how to do this.
Any help would be appreciated.
Thanks.
- j
Hi Justin
I'm interested in this as well. I discovered today that a project I'm developing threw several StaleObject errors, after I got a report of discrepancies in views of the data.
Aside... I did not even bother writing code to catch the error, eventhough I went to the trouble to add lock_version fields to all my tables. (I even wrote a nice little piece of code that reads the schema and creates lock_versions for all my tables. Got carried away I guess :^) ...Back.
I'm wondering how the data got stale. I'm certain the customer was the only person updating the rows. However, I have a piece of code that could be better written...
In the loop below, packing_item belongs_to rma and I do a separate save on each table record. The error occurred on the second save. Can this happen if only one person is running the application?
packing_slip.packing_items.each_with_index do |item, i| item.rma.state_id = PACKED
item.rma.packing_slip_id = packing_slip.id
item.rma.tracking_number = tracking[item.rma.number] [:tracking_number]
item.tracking_number = tracking[item.rma.number] [:tracking_number]
# save packing_item item.save!
# save rma <-- the error happened at this point. item.rma.save! end
Dave